Improving Optical Temperature Sensing Performance of Er3+ Doped Y2O3 Microtubes via Co-doping and Controlling Excitation Power

This work presents a new method to effectively improve the optical temperature behavior of Er3+ doped Y2O3 microtubes by co-doping of Tm3+ or Ho3+ ion and controlling excitation power. The influence of Tm3+ or Ho3+ ion on optical temperature behavior of Y2O3:Er3+ microtubes is investigated by analyzing the temperature and excitation power dependent emission spectra, thermal quenching ratios, fluorescence intensity ratios, and sensitivity. It is found that the thermal quenching of Y2O3:Er3+ microtubes is inhibited by co-doping with Tm3+ or Ho3+ ion, moreover the maximum sensitivity value based on the thermal coupled 4S3/2/2H11/2 levels is enhanced greatly and shifts to the high temperature range, while the maximum sensitivity based on 4F9/2(1)/4F9/2(2) levels shifts to the low temperature range and greatly increases. The sensitivity values are dependent on the excitation power, and reach two maximum values of 0.0529/K at 24 K and 0.0057/K at 457 K for the Y2O3:1%Er3+, 0.5%Ho3+ at 121 mW/mm2 excitation power, which makes optical temperature measurement in wide temperature range possible. The mechanism of changing the sensitivity upon different excitation densities is discussed.

Phosphors with high thermal stability are excellent candidate materials to be used for optical temperature sensing. Compared with the fluorides, Y 2 O 3 has the advantages of a high melting point, wide bandgap, high solubility between Y 3+ and Er 3+ , and good transparency in ultraviolet and infrared range 27 . Based on the green emissions from thermally coupled energy levels of 2 H 11/2 and 4 S 3/2 , the optical temperature sensing was studied in Er 3+ -Yb 3+ co-doped Y 2 O 3 bulks and nano-materials [27][28][29][30][31][32] . However, the aforementioned works have not dealt with excitation powers dependent optical temperature sensitivity, and lack to explore how to improve the optical temperature behaviors. The optical temperature property of Er 3+ doped phosphors was determined by the population of the thermally coupled energy levels of 2 H 11/2 and 4 S 3/2 12 . If the emission intensity ratios from the 2 H 11/2 and 4 S 3/2 is adjusted, the temperature sensitivity will change greatly. As reported that the energy gaps between 3 F 4 and 3 H 6 of Tm 3+ ion, as well as between 5 I 7 and 5 I 8 of Ho 3+ ion, were equal to the energy gap between 2 H 11/2 and 4 F 9/2 of Er 3+ ion 19 . Thus, the population of 2 H 11/2 and 4 S 3/2 levels of Er 3+ can be adjusted through the cross relaxation energy transfer between Er 3+ and Tm 3+ (or Ho 3+ ). In this case the optical temperature sensitivity will be easy controlled through controlling the concentration of the Tm 3+ or Ho 3+ ion. Considering the structure of energy levels of Tm 3+ and Ho 3+ , in this work, we propose a method to improve the optical temperature behavior of Er 3+ doped Y 2 O 3 through adjusting the green emission ratios with the cross-relaxation energy transfer between Er 3+ and Tm 3+ (or Ho 3+ ). The Er 3+ doped, Er 3+ -Tm 3+ co-doped, and Er 3+ -Ho 3+ co-doped Y 2 O 3 microtubes are synthesized, and their optical temperature behaviors are studied by controlling the excitation power of the laser and the concentration of Tm 3+ or Ho 3+ ions. It is observed that the optical temperature sensitivity of Er 3+ -doped Y 2 O 3 microtubes is significantly improved compared to the optical temperature sensitivity of Er 3+ -Yb 3+ co-doped Y 2 O 3 bulks and nano-materials.

Results
The low and high-magnification SEM images of Y 2 O 3 :1%Er 3+ , 1.5%Tm 3+ in Fig. 1(a) and (b) show that the feature of the sample is a hollow and tubular column, and the average diameter is 0.7μm. The EDS spectrum in Fig. 1(c) shows that the samples consist of O, Er, Tm, and Y, which is in good accordance with the initial elements in precursor solution. All the synthesized Y 2 O 3 :1%Er 3+ and Y 2 O 3 :1%Er 3+ , 0.5%Ho 3+ microtubes were analyzed also by the SEM and EDS. It is found that the shapes of all the samples are similar, and their sizes have no obvious change. Figure 2 displays the powders XRD patterns of Y 2 O 3 :1%Er 3+ , 1.5%Tm 3+ and Y 2 O 3 :1%Er 3+ , 0.5%Ho 3+ microtubes. The position and relative intensity of all the diffraction peaks can be readily indexed to the pure cubic Y 2 O 3 according to the JCPDS file no. 71-0099. No peak is recorded from other phases or impurities, which indicates the phase of the sample was pure Y 2 O 3 tubes. From XRD patterns, one can find that the Y 2 O 3 microtubes grow along the single direction (222) plane.
In order to study the role of Tm 3+ and Ho 3+ ions, the up-conversion spectra, red to green intensity ratios and CIE (X, Y) chromaticity coordinates of Y 2 O 3 :1%Er 3+ co-doped Tm 3+ and Ho 3+ are given in Fig. 3. Figure 3(a) shows the seven emission bands centered at 524 nm, 537 nm, 552 nm and 660 nm, 680 nm, 812 nm, 843 nm, which corresponds to the 2 H 11/2 → 4 I 15/2 (524-537 nm), 4 S 3/2 → 4 I 15/2 (552 nm), 4 F 9/2 → 4 I 15/2 (660-680 nm), and 4 I 9/2 → 4 I 15/2 (843 nm) transitions of Er 3+ ions and 3 H 4 → 3 H 6 (812 nm) transition of Tm 3+ ions. It is observed that the luminescence intensity of the green and 840 nm infrared emission decrease with increasing Tm 3+ concentration, and the intensity of red emission first increases and then decreases. The intensity decrease of the green and red emissions is attributed to the quenching induced by the cross relaxation energy transfer between Tm 3+ and Er 3+ : 4 F 7/2 + 3 H 6 → 4 I 9/2 + 3 H 5 31 . Figure 3(b) shows the red-green intensity ratio increases with the increase of Tm 3+ concentration, and luminescent color is tunable from green to red with the increase of Tm 3+ concentration. Similar spectrum modulation and color adjustment can be achieved through co-doping Ho 3+ into Y 2 O 3 :1%Er 3+ microtubes, as shown in Fig. 3(c) and (d). The intensities of the green and red emissions show the irregularly change with increasing Ho 3+ concentration, while the intensity of infrared emission has no obvious change. The intensity increase of the green and red emissions of Y 2 O 3 :1%Er 3+ , when the Ho 3+ concentration is  more than 1%, is attributed to the contribution of Ho 3+ on green and red emissions through the 5 S 2 → 5 I 8 (green) and 5 F 5 → 5 I 8 (red) transitions 33 . The Ho 3+ doping induces the increase of the red-green intensity ratio, and the enhancement of the yellow color, as shown in Fig. 3(d). After doping with Tm 3+ and Ho 3+ , the emission intensity, the red-green intensity ratio, and the color of Y 2 O 3 :Er 3+ are adjusted efficiently. This means that it is possible to adjust the optical temperature behavior of Y 2 O 3 :Er 3+ at a high temperature through doping Tm 3+ and Ho 3+ ions.
Temperature-dependent emission spectra of Y 2 O 3 :1%Er 3+ is shown in Fig. 4(a). It can be seen that the red and green emissions continuously decrease by increasing the temperature from 298 K to 573 K, without changing the peak positions of the emissions. The color shifting from yellow to green in Fig. 4(b) indicates the inhomogeneous decrease of green and red emissions induced by the high temperature. It is necessary to study the dependence of the fluorescence intensity ratio (R) on the temperature for the adjacent emission bands. The relation between R and T is expressed as: where a is constant, b is a correction term for the comprehensive population of thermally coupled energy levels induced by the thermal population, nonradiative relaxation and so on 2, 34 . Relative sensitivity is one of the key parameters to determine the suitability for optical thermometry, and is defined as where a and b are the constants from Eq. (1). Figure 4(c) shows temperature-dependent emission intensity ratios of several adjacent emission bands. The experimental data points can be fitted well with a line model. The slope values of fitted lines are dependent on the combination types of adjacent emission bands. It means that the sensitivity values are different when we use the different adjacent emission bands as the thermal coupled levels. Figure 4(d) shows the temperature dependent sensitivity values of five thermal coupled levels. One can find that all the sensitivity values increase and then decrease with the temperature increase, exhibiting the maxima values at different temperature points. The maximum values at (101 K, 0.0050/K), (461 K, 0.0027/K), (360 K, 0.0018/K), (413 K, 0.0044/K), (52 K, 0.0142/K) are observed for the adjacent thermal coupled levels of 524 nm/537 nm, 524 nm/552 nm, 537 nm/552 nm, (524 + 537) nm/552 nm, 660 nm/680 nm. Notably, the adjacent thermal coupled levels of (524 + 537) nm/552 nm shows the large intensity in the high temperature range. The adjacent thermal coupled levels of 660 nm/680 nm shows the very large intensity in the low temperature range. As reported, the Er 3+ -Yb 3+ co-doped Y 2 O 3 sphere nano-particles showed the maximum sensitivity value of 0.0044/K at 427 K 28 , and the Er 3+ -Yb 3+ -Eu 3+ tri-doped Y 2 O 3 sphere nanoparticles showed the maximum sensitivity value of 0.0103/K at 593 K 27 . In contrast, our Er 3+ -doped Y 2 O 3 microtubes show a large sensitivity value of 0.0142/K in the low temperature range, and an excellent sensitivity value of 0.0044/K in the high temperature range. Most of the sensors based on up-conversion luminescence of Er 3+ ion showed excellent sensitivity properties at the high temperature of more than 300 K 12, 14 , while it was reported rarely on optical thermometry below room temperature.
The influence of Tm 3+ and Ho 3+ on the optical temperature behaviors of Y 2 O 3 :1%Er 3+ is studied through co-doping 0.2 mol% Tm 3+ and 0.5 mol% Ho 3+ into Y 2 O 3 :1%Er 3+ , as shown in Figs 5 and 6. Compared with the Fig. 4, it is evident that the emission spectra, CIE chromaticity coordinates, and R values of Y 2 O 3 :1%Er 3+ are adjusted after co-doping with Tm 3+ or Ho 3+ . Importantly, after co-doping with Tm 3+ and Ho 3+ ions, the maximum sensitivity value based on the thermal coupled levels of (524 + 537) nm/552 nm, is greatly enhanced and shifts to the higher temperature range, while the maximum sensitivity value based on the thermal coupled levels of 660 nm/680 nm, shifts to lower temperature range and increases a lot. This makes it possible to achieve the optical temperature measurement in the low temperature range.
It is necessary to study the thermal stability of thermal coupled levels in the process of optical temperature sensing. The thermal stability of emission bands can be determined by the number change of photons involved in the up-conversion processes at a different temperature. The up-conversion emission intensity I and excitation power P is expressed as follows: where I is the emission intensity, P is incident pump power, and n is the number of pump photons absorbed in the up-conversion process 35 . Figure 7 shows the double logarithmic plots of the emission intensity I as a function of pump power P of the Y 2 O 3 :Er 3+ . The fit results indicate that three infrared photons are needed to emit green and red luminescence at 298 K and 573 K. After doping Tm 3+ and Ho 3+ into Y 2 O 3 :Er 3+ , the fit results indicate that two infrared photons are needed to emit green and red luminescence at 298 K and 573 K, as shown in Figs 8 and 9. The decrease of slope values for green and red emissions means that the up-conversion process becomes easy to occur at the same excitation power. Thus, the thermal stability of Y 2 O 3 :Er 3+ microtubes is improved through co-doping Tm 3+ and Ho 3+ ions. M. Pollnau observed that the population processes of green and red emissions were adjusted by the excitation power density 35 . It is necessary to study the excitation power-dependent optical temperature behaviors of Y 2 O 3 :1%Er 3+ , Y 2 O 3 :1%Er 3+ , 0.2%Tm 3+ , and Y 2 O 3 :1%Er 3+ , 0.5%Ho 3+ . To evaluate the influence of excitation powers on luminescence quenching, the thermal quenching ratio (TRQ) of Er 3+ emission bands at different excitation powers are studied in Figs 10, 11 and 12. The thermal quenching ratio (TRQ) of emission bands induced by the temperature change is defined as follows where I T is luminescence intensity at a different temperature T, and I R is luminescence intensity at room temperature 36 . From Fig. 10(a) and (b), it is obvious that the TRQ of green and red emissions of Y 2 O 3 :1%Er 3+ is strongly dependent on temperature and excitation powers. The values of TRQ of green and red emissions increase with the increase of temperature, change irregularly with the increase of excitation powers. The red-to-green intensity ratio is dependent on both temperature and excitation powers, and shows the large values at the excitation power of 121 mW/mm 2 , as shown in Fig. 10(c). After co-doping with the Tm 3+ and Ho 3+ , the TRQ values of green and red emissions and red-to-green emission intensity ratios of Y 2 O 3 : 1%Er 3+ are adjusted, as shown in Figs 11 and 12.
The influence of excitation powers on the sensitivity values for the Y 2 O 3 :1%Er 3+ are studied in Fig. 13. At different excitation powers, based on the ratio of (524 + 537) nm/552 nm, the maximum sensitivity value of 0.0049/K  at 471 K is achieved at 30 mW/mm 2 , as shown in Fig. 13(a). Based on the ratio of 660 nm/680 nm, the maximum sensitivity value of 0.0382/K at 34 K is achieved at 322 mW/mm 2 , as shown in Fig. 13(b). After co-doping with Tm 3+ , the maximum sensitivity values are significantly enhanced and shift to the high temperature range with various Tm 3+ concentrations, as shown in Fig. 14(a) and (b). The sensitivity value based on the ratio of (524 + 537) nm/552 nm achieves the maximum value at (459 K, 0.0055/K) when the Tm 3+ concentration is 0.2 mol%. The sensitivity value based on the ratio of 660 nm/680 nm achieve the maximum value at (37 K, 0.0235/K) when the Tm 3+ concentration is 1.0 mol%. Furthermore, the excitation power dependent sensitivity is studied in Fig. 14(c) and (d). The maximum sensitivity values change irregularly with the increase of the excitation powers, and reach the maximum value of (504 K, 0.0056/K) at 30 mW/mm 2 for the ratio of (524 + 537) nm/552 nm, and the maximum value of (22 K, 0.0282/K) at 85 mW/mm 2 for the ratio of 660 nm/680 nm. Similarly, the influence of Ho 3+ concentration and excitation powers on the sensitivity values are also studied in Fig. 15. The optimized Ho 3+ concentration is obtained is 0.5 mol%. For the Y 2 O 3 :1%Er 3+ , 0.5%Ho 3+ , the maximum sensitivity values change irregularly with the increase of the excitation powers, and reach the maximum value of (457 K, 0.0057/K) at 184 mW/mm 2 for the ratio of (524 + 537) nm/552 nm, and the maximum value of (24 K, 0.0529/K) at 121 mW/mm 2 for the ratio of 660 nm/680 nm. Compared with the Y 2 O 3 :1%Er 3+ , the maximum sensitivity value increases from 0.0049/K (0.0382/K) to 0.0057/K (0.0529/K) through co-doping with the 0.5 mol% Ho 3+ ions. In order to compare the sensitivity values of Y 2 O 3 doped with other ions for the fluorescence thermometric study, the reported sensitivity values based on the up-conversion fluorescence of Er 3+ are listed in Table S1 in the supplementary information. It can be seen that the sensitivity values in this paper are higher than that of other Y 2 O 3 materials not only in the high temperature range but also in the low temperature range. It means that it is a good method to improve the sensitivity of Y 2 O 3 :Er 3+ through co-doping Ho 3+ ions.
To study the influence of doping concentration and excitation powers on the optical temperature behaviors, the dynamic balance rate-equation models for the energy transfer between Er 3+ and Tm 3+ (or Ho 3+ ) are established in Figs S1 and S2 in the supplementary information. The population dynamic process of excited states is simulated by using the eight-level model. The population density of the 2 H 11/2 / 4 S 3/2 and 4 F 9/2(1) / 4 F 9/2(2) thermal coupled levels can be obtained in the Equations S(8), S(9), S(17), and S (18). It is obvious that the values of N 3 and N 4 are dependent on not only the excitation power (ρ) and doping concentration (N 0 , N 6 ), but also the cross relaxation rates (W c1 , W c2 , W c3 , and W c4 ), nonradiative decay rates (W ij ), and radiative transition rates (A ij ). Notably, the cross relaxation rates, nonradiative decay rates, and radiative transition rates are strongly dependent on the temperature 37 . It means that the population processes of the 2 H 11/2 / 4 S 3/2 and 4 F 9/2(1) / 4 F 9/2(2) thermal coupled levels are determined by the excitation power, doping concentration and temperature. It is a complex multi-field coupling effect. Thus, the sensitivity shows irregularly change with the increase of excitation powers and doping concentrations.
Briefly speaking, at low excitation power, the thermal coupled levels of 4 S 3/2 and 2 H 11/2 are populated by the ground state absorption (GSA) and excited state absorption (ESA), and then the multiphonon nonradiative relaxation (NR) from 4 F 7/2 level, shown in Fig. S1. The thermal coupled levels of 4 F 9/2(1) and 4 F 9/2(2) are populated by the NR process from the 4 S 3/2 level. According to the Boltzmann distribution 14,15 , the population of the 2 H 11/2 level  increases with respect to the 4 S 3/2 with the increase of the temperature, owe to the low energy difference between 4 S 3/2 and 2 H 11/2 levels. The thermalization of the 2 H 11/2 level is dominant and the depopulation induced by the NR process can be neglected. At high excitation power, the population saturation effect of the 4 S 3/2 and 2 H 11/2 levels can be observed 35,38 . Therefore, the population of 2 H 11/2 level changes a little with the temperature. The NR is easy to occur in the case of small energy difference between adjacent energy levels at the high temperature 19,20 . The NR process is dominant to depopulate the 2 H 11/2 level, due to the small energy difference between 2 H 11/2 and 4 S 3/2 (ΔE = 968 cm −1 ). Thus, the R in equation 1 decreases with the increase of excitation power density, due to  the fact that the NR possibility increases with the temperature increase. As a result, the sensitivity decreases at high excitation power density. Importantly, after co-doping Tm 3+ and Ho 3+ ion, the sensitivity values shift to higher and lower temperature ranges, and are greatly increased. It is attributed to the fact that the populations of 4 S 3/2 / 2 H 11/2 and 4 F 9/2(1) / 4 F 9/2(2) levels are adjusted by the cross relaxation process, such as CR1 and CR3. With the concentration further increase of the Tm 3+ and Ho 3+ ions, the green and red emissions are quenched, due to the cross relaxation process, such as CR2 and CR4.

Conclusions
In summary, we explore a method to improve the photoluminescence and optical temperature sensing of Er 3+ doped Y 2 O 3 microtubes through combining the ion doping with the control of excitation powers. It is found that the optical temperature behaviors of Er 3+ doped Y 2 O 3 microtubes are strongly dependent on the ion doping and excitation powers. After doping Tm 3+ or Ho 3+ ions, the spectrum of Er 3+ doped Y 2 O 3 microtubes is modified, the thermal quenching behavior of Y 2 O 3 :Er 3+ microtubes is inhibited, and the optical temperature sensitivity is significantly enhanced. It is achieved that the maximum sensitivity value is 0.0529/K at low temperature range of less than 250 K while it is 0.0057/K at more than 250 K by controlling the excitation power to 121 mW/mm 2 . The maximum sensitivity value of 0.0529/K at 24 K is superior to that of our earlier report. It makes up the lack of optical temperature detection in the low temperature range below 250 K.
The Re 2 O 3 (Re = Er, Y, Tm, and Ho) was dissolved in hydrochloric acid, and then the solution was heated to evaporate the water completely. The obtained rare earth metal trichloride (ReCl 3 ) was dissolved in deionized water to prepare the solutions of ReCl 3 (0.2 mol L −1 ). Er 3+ doped Y 2 O 3 microtubes were prepared by a hydrothermal method. In a representative synthesis process, an aqueous solution of 9.90 mL YCl 3 (0.2 mol L −1 ), 0.10 mL ErCl 3 (0.2 mol L −1 ) was mixed with 28 mL of distilled water under thorough stirring. It was then vigorously stirred by a magnetic stirrer at room temperature for 30 min, while 3.5 mL of a 5 mol L −1 NaOH solution was slowly added in drops. The colloidal solution was then transferred into an Teflon vessel at 473 K for 24 h. The final products were collected, washed several times with ethanol, and purified by centrifugation. Samples were then dried in an oven for 6 h at 373 K. Finally, the dried powders were sintered at 1173 K for 3 h in an electric annealing furnace. After then, the samples were cooled down rapidly to room temperature. The same method was used for Er 3+ / Tm 3+ and Er 3+ /Ho 3+ co-doped Y 2 O 3 microtubes.
The structure of the sample was investigated by X-ray diffraction (XRD) using X'TRA (Switzerland ARL) equipment provided with Cu tube with Kα radiation at 1.54056 Å. The size and shape of the sample were observed by a JSM-IT300 scanning electron microscope (SEM) (JEOL Ltd., Tokyo, Japan) equipped with an energy dispersive X-ray spectrometer (EDS). Luminescence spectra were obtained by the Acton SpectraPro Sp-2300 Spectrophotometer with a photomultiplier tube equipped with a 980 nm laser as the excitation source. Different temperature spectra were obtained in the range 298-573 K by using an INTEC HCS302 Hot and Cold System.